Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    providing understanding

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     attack
    0.42
     offence
    0.40
     ataque
    0.40
     notificación
    0.39
     reception
    0.39
     aptitude
    0.39
     satisfaction
    0.39
     repercussions
    0.38
     authorisation
    0.38
     competente
    0.38
    POSITIVE LOGITS
    ~!
    0.45
    鋳
    0.44
    !”.
    0.43
     !”
    0.43
     কিভাবে
    0.41
     『
    0.40
    !).
    0.39
    ！」
    0.39
    Tela
    0.39
    すぎる
    0.38
    Act Density 0.002%

    No Known Activations